@techreport{oai:ipsj.ixsq.nii.ac.jp:00218025,
 author = {Shota, Yamada and Rizk, Hamada and Hirozumi, Yamaguchi and Shota, Yamada and Rizk, Hamada and Hirozumi, Yamaguchi},
 issue = {38},
 month = {May},
 note = {The demand for safety-enhancing solutions is on the rise, especially due to COVID-19's rapid spread. In order to track infected cases and hence restrict the spread of the virus, real-time life-logging is an essential application. This application highlights the necessity for a precise human identification technique in situations when cameras are not feasible owing to privacy concerns. The potential of the LiDAR sensor to represent the surrounding world in the form of a 3D point cloud has recently gained interest. In this paper, we present a new wearable device with a small-sized LiDAR that may be used to create an onboard human identification system for life-logging. Our proposed system starts with clustering to remove noise and background. Then fisher features are extracted from them. After that, the collected characteristics are utilized to train classifiers to identify the subjects. We conducted two different experiments to evaluate the suggested system. We collected six and thirteen subjects for each experiment. The results show that the proposed system can effectively remove noise and accurately identify subjects with at least 95% accuracy in both experiments., The demand for safety-enhancing solutions is on the rise, especially due to COVID-19's rapid spread. In order to track infected cases and hence restrict the spread of the virus, real-time life-logging is an essential application. This application highlights the necessity for a precise human identification technique in situations when cameras are not feasible owing to privacy concerns. The potential of the LiDAR sensor to represent the surrounding world in the form of a 3D point cloud has recently gained interest. In this paper, we present a new wearable device with a small-sized LiDAR that may be used to create an onboard human identification system for life-logging. Our proposed system starts with clustering to remove noise and background. Then fisher features are extracted from them. After that, the collected characteristics are utilized to train classifiers to identify the subjects. We conducted two different experiments to evaluate the suggested system. We collected six and thirteen subjects for each experiment. The results show that the proposed system can effectively remove noise and accurately identify subjects with at least 95% accuracy in both experiments.},
 title = {Human Identification Based On Point Cloud Captured By Small-Size LiDAR},
 year = {2022}
}